Diversified Trend Trading Approach

super

better mar
in my opinion is all about the mar?

added one more strategy + deleted agressive entries.
impact is dramatic..well, try to improve mar by 0,05

Exactly J. It is all about the MAR.

The strategies we deploy are simply the tools to derive the optimal path that leads to the best compounded returns. Compounding over a time series as opposed to the strategies themselves is actually what does the 'heavy lifting' in terms of overall performance.

The path of the return stream (or it's adverse downside volatility) is what compromises the compounding effect of a return stream. In fact adverse drawdowns early on in the overall path significantly effect the final net wealth at the end of the series. The step-ups on the other hand (favourable upside volatility) are actually beneficial for compounding particularly if they occur earlier in the series. Having a strategy that has unbounded upside potential for the equity curve is very favorable for compounding.

MAR (defined as CAGR/Max Draw for the entire series) is the best performance measure (better than Calmar) to lead us to the optimal compounded return stream.

For a start MAR uses the Compound Annual Growth Rate in its calculation that takes compounding into account and also explicitly highlights the maximum adverse excursion in % returns of the series. MAR as opposed to Calmar operates over the entire time series. MAR is a far superior performance metric than say the Sharpe or Sortino ration which fails to take compounding (geometric returns) into account.

For example if you plot monthly percentage returns of a return stream you can find instances where total summated monthly % returns are greater than another return series....yet the inferior summated monthly return series generates a superior net wealth result. This is because compounding is incrementally applied over a time series and a major drawdown early in the series significantly compounds the non-linear incremental compounding effect.

Below is a classic example that highlights this effect.



The summated total returns form the 20 year period of the TF Index is only 160.83% over the time series. The summated total returns of the S&P500TR Index is 166.19% for the entire period. So one would assume the performance of the S&P500 TR series is superior when in fact it is not.

Note how the CAGR of the 2 different return streams of 7.28% for the TF Index and 6.33% for the S&P picks up this performance discrepancy. The CAGR correctly reflects the performance difference as opposed to the total summated % returns which is the 'illusion'.

The total equity for the TF Index at the end of the series is $4.27 versus $3.56 when compounded monthly. This is a huge difference in end result between these comparisons of 34% despite the TF Index possessing lower total summated returns than the S&P500TR Index.

Most traders who use inferior performance metrics apart from MAR simply cannot fathom this and would be attracted to the S&P Index over the TF Index every time given the perceived difference in performance return....yet they would be VERY wrong.

The secret lies in mitigating adverse returns (refer to the S&P table and the monthly columns highlighted in red). The more extreme adverse returns (left tailed event) of a buy and hold in the S&P500TR series are avoided by the TF Index (refer to TF Table) that cut losses short and never allow adverse excursions to occur in the left tail but allow for right tailed outliers and why the compounded series performs better over the long term.

This is WHY diversified systematic trend followers cut losses short and let profits run at all times and trade both long or short. It ensures we produce a superior risk-adjusted return to alternatives....which is another way of saying that our solutions produce the optimal path for compounding.

Those portfolio managers who do not deploy trend following strategies and are unidirectional in their posture (either long or short) such as long only strategies.....need to therefore apply 'Tail Risk' hedges to get them on the optimal path for compounding. But in our simplified trend following models that always cut losses short but let profits run already have this 'tail risk' hedge built into them through design. 'Tail Risk' supplemental add-ons are never as efficient as fully integrated risk-adjusted approaches.

Which offers greatest scope to the expectancy equation....the win rate or the reward to risk relationship?

For the trend follower that accepts that most liquid instruments can at times provide fat tailed returns through offering extreme profit potential .........then it is always preferable to strike for a higher Reward to Risk relationship by applying a trailing stop with no profit target than by attempting to improve the expectancy result by lifting the win rate by using a profit target.

The win rate is expressed as a % relationship that lies somewhere between 0 and 100%. This rate relates to the number of successful events that are profitable. It is silent in relation to whether the wins are materially significant or not. Now in a Non-Gaussian world where fat tails reside, a single event can lead to a profit that is 'extreme' so we are less concerned with the number of successful events that can be materially insignificant to overall expectancy and more concerned with unique events of unbridled magnitude that materially impact the expectancy value. Win rate is therefore far less important than the significance of material events for a diversified trend follower.

So we can participate in outlier events if we have an understanding of where the Gaussian distribution ends and the Non Gaussian components of the tails of the distribution commences and then...... coupled with extensive market and trend following system diversification....only participate in the market when these more extreme conditions arise.

This way we avoid overtrading when market conditions are not extreme yet from extreme diversification can still lift trade frequency of more exotic events combined with our win rate. Diversification plus only trading during fat tailed regimes while cutting all losses short allows us to achieve superior risk adjusted returns to alternative approaches.

The actual wealth delivered by the infrequent and unpredictable tail events is therefore not compromised by the inevitable whipsaws that accompany normal trends that commence within the noise or mean reverting environment...... and allows us to only focus on trends that extend into the zones of 'outlier' potential.

Of course we can only attempt to achieve this response if we are fully systematic in our application and in our method of quantitative analysis. The suite of diversified methods applied to trading in Non Gaussian regimes needs to be extensive to capture these rare and unpredictable opportunities and we need to use the data itself as a method of identifying when price lies outside normal Gaussian boundaries.

You see to be a participant in a wealth event of unlimited magnitude you need to be 'in the right place' at the 'right time'. This is not a mere 'statement of luck' as many assume... but is a circumstance that can be addressed by:

1. capital preservation (knowing when to trade and when not to trade so you can be eligible to participate in the unpredictable event);
2. systematic processes (so you can participate at any time that the market dictates 24/5);
3. diversification (aka being in the right place);
4. quantitative techniques.....aka knowing when to participate in the extreme event (aka the right time).

CTA Fund Performance Report - 30 September 2020 - Controlling Risk to Optimise the Path to Wealth

Attached Image (click to enlarge)

https://atstradingsolutions.com/cta-...ath-to-wealth/

The foundations of Trend Following (TF) and Natural Selection (NS) are based on survival. From this foundation NS amplifies survivability through reproduction....whereas TF amplifies survivability through diversification allowing a trader to participate in beneficial outliers.

Unlike predictive methods...TF is a blind forecasting process that simply capitalises on how markets are understood to behave. The focus is 100% directed towards managing adverse risk to enable the participant to survive for extended periods....of hopefully sufficient extent........ to then be in a position to occasionally enjoy the bounty that the market unpredictably delivers from time to time.

Cut losses short at all times...but let profits run with no profit targets.

Your research conducted through backtests is not to determine profitability as this is unpredictable in nature and unquantifiable due to the absence of profit targets. All your effort in research is in determining 'how to cut losses short'. (Eg. what is too short....versus what is too wide.)

Your backtest is simply a method to identify 'how to break your model'. It is designed to identify risk weaknesses that must be addressed based on historic market conditions. Obviously the longer the backtest the better..... as this is a proxy to ensure that your model has been assessed under as great a variety of different market condition as you can find.

Having a robust backtest simply is a statement about likely 'survivability' or robustness....and is not a statement about future profitability under uncertainty.

Part 1 of a Video Series that Demonstrates how to Develop Diversified Systematic Trend Following Models using Strategy Quant X and Quant Analyzer.

Enjoy :-)

The Power of Compounding. Why wealth building is all about robustness and patience.

You made it sound so interesting and yes, the system is also great with low risk. It will be better if you share some details.

I will be rolling out further episodes shortly that describe how we generate these systematic portfolios using SQX so stay tuned. :-)

Part 2 of a Video Series that Demonstrates how to Develop Diversified Systematic Trend Following Models using Strategy Quant X and Quant Analyzer.

Part 3 of a Video Series that Demonstrates how to Develop Diversified Systematic Trend Following Models using Strategy Quant X and Quant Analyzer.

Part 4 of a Video Series that Demonstrates how to Develop Diversified Systematic Trend Following Models using Strategy Quant X and Quant Analyzer.

Part 5 of a Video Series that Demonstrates how to Develop Diversified Systematic Trend Following Models using Strategy Quant X and Quant Analyzer.

Part 6 of a Video Series that Demonstrates how to Develop Diversified Systematic Trend Following Models using Strategy Quant X and Quant Analyzer.

Part 7 of a Video Series and Finale that Demonstrates how to Develop Diversified Systematic Trend Following Models using Strategy Quant X and Quant Analyzer.

Inserted Video

VIDEO SERIES – HOW TO DEVELOP DIVERSIFIED SYSTEMATIC TREND FOLLOWING MODELS FOR METATRADER 4 USING STRATEGYQUANT X AND QUANTANALYZER.

I have compiled all 7 episodes of the entire series in this single Blog Post.


Enjoy

C

CTA Fund Performance Report - 31 October 2020 - It is Better to Win Big than Win Often

Attached Image

https://atstradingsolutions.com/cta-...han-win-often/

There's gold in them thar hills